Aditya Raushan

Key Skills

Software

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Text

Top Label Types

RLHF

Fine Tuning

Evaluation Rating

Freelancer Overview

I have hands-on experience in AI training data and model optimization, particularly through my freelance work as an LLM Trainer, where I fine-tuned RLHF models to enhance performance, accuracy, and domain-specific language understanding. My technical skills include Python, C++, data structures, and algorithms, which I have applied to both software engineering and AI tasks. I am proficient in unit testing, data annotation, and leveraging efficient algorithms for large-scale data processing. My projects span domains such as medical inventory management and intelligent route planning, where I have worked extensively with databases like MySQL and MongoDB, as well as data visualization tools. I am passionate about building reliable AI systems and ensuring high-quality, well-labeled datasets for robust model training.

Labeling Experience

Source Code Annotation, Evaluation, and AI Training Data Preparation for Programming Models

Scale AITextRLHFFine Tuning

Worked on programming-related data labeling and evaluation tasks to support training and improvement of AI code generation models. Responsibilities included reviewing AI-generated code outputs, validating correctness, and annotating responses based on quality, accuracy, and instruction compliance. Evaluated code across multiple programming languages including Java, Python, and JavaScript. Performed tasks such as: Comparing multiple code solutions for correctness and efficiency Identifying logical errors, syntax issues, and edge case failures Writing high-quality reference solutions for model training Creating prompt-response pairs for supervised fine-tuning datasets Ensuring outputs followed formatting, compilation, and execution requirements Worked with structured evaluation rubrics focusing on: Code correctness Readability and maintainability Time and space complexity awareness Instruction-following accuracy Edge case handling Maintained high annotation quality by follo

Worked on programming-related data labeling and evaluation tasks to support training and improvement of AI code generation models. Responsibilities included reviewing AI-generated code outputs, validating correctness, and annotating responses based on quality, accuracy, and instruction compliance. Evaluated code across multiple programming languages including Java, Python, and JavaScript. Performed tasks such as: Comparing multiple code solutions for correctness and efficiency Identifying logical errors, syntax issues, and edge case failures Writing high-quality reference solutions for model training Creating prompt-response pairs for supervised fine-tuning datasets Ensuring outputs followed formatting, compilation, and execution requirements Worked with structured evaluation rubrics focusing on: Code correctness Readability and maintainability Time and space complexity awareness Instruction-following accuracy Edge case handling Maintained high annotation quality by follo

2025 - 2025

Education

Work History